SEARCH · Engineering Papers

Results for “train”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 523 records · Page 29

All sources NASA NTRS DOE OSTI

NASA NTRS2016

Clinical Outcome Metrics for Optimization of Robust Training

Introduction: The emphasis of this research is on the Human Research Program (HRP) Exploration Medical Capability's (ExMC) "Risk of Unacceptable Health and Mission Outcomes Due to Limitations of In-Flight Medical Capabilities." Specifically, this project aims to contribute to the closure of gap ExMC 2.02: We do not know how the inclusion of a physician crew medical officer quantitatively impacts clinical outcomes during exploration missions. The experiments are specifically designed to address clinical outcome differences between physician and non-physician cohorts in both near-term and longer-term (mission impacting) outcomes. Methods: Medical simulations will systematically compare success of individual diagnostic and therapeutic procedure simulations performed by physician and non-physician crew medical officer (CMO) analogs using clearly defined short-term (individual procedure) outcome metrics. In the subsequent step of the project, the procedure simulation outcomes will be used as input to a modified version of the NASA Integrated Medical Model (IMM) to analyze the effect of the outcome (degree of success) of individual procedures (including successful, imperfectly performed, and failed procedures) on overall long-term clinical outcomes and the consequent mission impacts. The procedures to be simulated are endotracheal intubation, fundoscopic examination, kidney/urinary ultrasound, ultrasound-guided intravenous catheter insertion, and a differential diagnosis exercise. Multiple assessment techniques will be used, centered on medical procedure simulation studies occurring at 3, 6, and 12 months after initial training (as depicted in the following flow diagram of the experiment design). Discussion: Analysis of procedure outcomes in the physician and non-physician groups and their subsets (tested at different elapsed times post training) will allow the team to 1) define differences between physician and non-physician CMOs in terms of both procedure performance (pre-IMM analysis) and overall mitigation of the mission medical impact (IMM analysis); 2) refine the procedure outcome and clinical outcome metrics themselves; 3) refine or develop innovative medical training products and solutions to maximize CMO performance; and 4) validate the methods and products of this experiment for operational use in the planning, execution, and quality assurance of the CMO training process The team has finalized training protocols and developed a software training/testing tool in collaboration with Butler Graphics (Detroit, MI). In addition to the "hands on" medical procedure modules, the software includes a differential diagnosis exercise (limited clinical decision support tool) to evaluate the diagnostic skills of participants. Human subject testing will occur over the next year.

Ebert, D.↗

NASA NTRS2015

SimSup's Loop: A Control Theory Approach to Spacecraft Operator Training

Immersive simulation is a staple of training for many complex system operators, including astronauts and ground operators of spacecraft. However, while much has been written about simulators, simulation facilities, and operator certification programs, the topic of how one develops simulation scenarios to train a spacecraft operator is relatively understated in the literature. In this paper, an approach is presented for using control theory as the basis for developing the immersive simulation scenarios for a spacecraft operator training program. The operator is effectively modeled as a high level controller of lower level hardware and software control loops that affect a select set of system state variables. Simulation scenarios are derived from a STAMP-based hazard analysis of the operator's high and low level control loops. The immersive simulation aspect of the overall training program is characterized by selecting a set of scenarios that expose the operator to the various inadequate control actions that stem from control flaws and inadequate control executions in the different sections of the typical control loop. Results from the application of this approach to the Lunar Atmosphere and Dust Environment Explorer (LADEE) mission are provided through an analysis of the simulation scenarios used for operator training and the actual anomalies that occurred during the mission. The simulation scenarios and inflight anomalies are mapped to specific control flaws and inadequate control executions in the different sections of the typical control loop to illustrate the characteristics of anomalies arising from the different sections of the typical control loop (and why it is important for operators to have exposure to these characteristics). Additionally, similarities between the simulation scenarios and inflight anomalies are highlighted to make the case that the simulation scenarios prepared the operators for the mission.

LADEE↗

NASA NTRS2017

A 3D Active Learning Application for NeMO-Net, the NASA Neural Multi-Modal Observation and Training Network for Global Coral Reef Assessment

NeMO-Net, the NASA neural multi-modal observation and training network for global coral reef assessment, is an open-source deep convolutional neural network and interactive active learning training software aiming to accurately assess the present and past dynamics of coral reef ecosystems through determination of percent living cover and morphology as well as mapping of spatial distribution. We present an interactive video game prototype for tablet and mobile devices where users interactively label morphology classifications over mm-scale 3D coral reef imagery captured using fluid lensing to create a dataset that will be used to train NeMO-Nets convolutional neural network. The application currently allows for users to classify preselected regions of coral in the Pacific and will be expanded to include additional regions captured using our NASA FluidCam instrument, presently the highest-resolution remote sensing benthic imaging technology capable of removing ocean wave distortion, as well as lower-resolution airborne remote sensing data from the ongoing NASA CORAL campaign. Active learning applications present a novel methodology for efficiently training large-scale Neural Networks wherein variances in identification can be rapidly mitigated against control data. NeMO-Net periodically checks users input against pre-classified coral imagery to gauge their accuracy and utilize in-game mechanics to provide classification training. Users actively communicate with a server and are requested to classify areas of coral for which other users had conflicting classifications and contribute their input to a larger database for ranking. In partnering with Mission Blue and IUCN, NeMO-Net leverages an international consortium of subject matter experts to classify areas of confusion identified by NeMO-Net and generate additional labels crucial for identifying decision boundary locations in coral reef assessment.

NeMO=NET↗

NASA NTRS2014

Transfer-of-Training Research at NASA Ames Research Center

In this presentation, I will first give an overview of some of the research being performed at the Human Systems Integration Division of NASAs Ames Research Center. In the second part of the presentation, I will discuss two studies that investigate the effects of simulator motion on pilot transfer of training. In a study funded by the Federal Aviation Administration, task performance is used to investigate the effects of motion on the training of 4 challenging tasks in a transport category aircraft. In addition, the effectiveness of recently proposed objective motion cueing criteria for training simulators are evaluated. The second study, part of the NASA Aeronautics Research Program, focuses on optimizing simulator motion for maximum transfer of stall recovery training and utilizes a cybernetic approach to measure transfer of training.

Zaal, Peter M. T.↗

NASA NTRS2014

Transfer of Training on the Vertical Motion Simulator

This paper describes a transfer-of-training study in the NASA Ames Vertical Motion Simulator (VMS). Sixty-one general aviation pilots, divided in four groups, trained on four challenging commercial transport tasks with four motion conditions: no motion, small hexapod, large hexapod, and VMS motion. Then, every pilot repeated the tasks with VMS motion to determine if training with different motion conditions had an effect. New objective motion criteria guided the selection of the motion parameters for the small and large hexapod conditions. Considering results that were statistically significant, or marginally, the motion condition used in training affected 1) longitudinal and lateral touchdown location; 2) the number of secondary stall warnings in a stall recovery; 3) pilot ratings of motion utility and maximum load factor obtained in an overbanked upset recovery; and 4) pilot ratings of motion utility and pedal input reaction time in the engine-out-on-takeoff task. Since the motion condition revealed statistical differences on objective measures in all the tasks, even with some in the direction not predicted, trainers should be cautious not to oversimplify the effects of platform motion. Evidence suggests that the new objective motion criteria may offer valid standardization benefits, as instances arose when the higher-fidelity hexapod motion, as predicted by the criteria, provided better cues in training than the lower-fidelity hexapod motion.

Zaal, Peter M. T.↗

NASA NTRS2020

Monitoring for Flight Path Management: Inputs to Pilot Monitoring Training

The commercial aviation industry world-wide has identified a need for improved pilot monitoring (e.g., ICAO, 2016). More specifically, aviation safety data indicate that failures in pilots' monitoring for flight path management (FPM) have contributed to a range of undesired outcomes: accidents, major upsets, and non-compliance with ATC guidance. The FAA has further stated that these types of FPM failures are likely to worsen with the increasingly complex air traffic control systems and FPM concepts proposed for NextGen (https://www.faa.gov/nextgen/what_is_nextgen/) operations (e.g., see Hah et al., 2017). An important element of this additional complexity will be the introduction of new automation or artificial intelligence that is intended to work with the flight crew but can add additional monitoring burdens. One potential mitigation for this situation is to enhance pilot training for effective monitoring. NASA Ames Research Center was asked to identify and evaluate training approaches that have the potential to enhance pilots' ability to effectively monitor for FPM (with the result of improved awareness). The focus of this work is to identify, develop or validate training guidance to improve pilot monitoring/awareness regarding FPM and mitigate the recent trend of accidents and incidents, especially Loss of Control (LOC) events. The result of this work should be guidance for FAA training and standards organizationsâ€”such as the Air Carrier Training Aviation Rulemaking Committee (ACT ARC)â€”to aid in reducing the risk of incidents and accidents due to inadequate pilot monitoring/awareness.

Feary, Mike↗

NASA NTRS2020

LUNAR SURFACE GEOSCIENCE TRAINING FOR ASTRONAUTS

Geoscientists have been training and preparing astronauts to observe the Earth from space and explore other planetary surfaces with a legacy that reaches back to the early days of the space program. Continuing this legacy and critical function, a core NASA team has been closely coordinating with the Johnson Space Center (JSC) Flight Operations Directorate (FOD) as well as academic, research institutions, and other governmental partners to conduct a comprehensive geoscience training program that ranges from initial astronaut candidate training to preparing assigned crew for future lunar surface missions. Described below are the three program training phases (summarized in Figure 1), along with recent program highlights and forward planning. Given NASA’s increased planning and development efforts regarding near-term lunar surface science activities, we welcome ideas, concepts, capabilities, collaborations, and innovations to this astronaut training program.

T. G. Graff↗

NASA NTRSDate not supplied

Training Astronauts using Hardware-in-the-Loop Simulations and Virtual Reality

The commercial market has recently started giving significant attention to virtual and augmented reality, even though the technology has been around for many years. The Virtual Reality Training Lab (VRL) at the NASA Johnson Space Center has been using virtual reality to train astronauts for decades. This paper will focus on describing three major Hardware-in-the-Loop VR simulation systems, the Simplified Aid for EVA Rescue (SAFER) system known as the "jetpack", the Mass Handling System nicknamed Charlotte, and a simulated robotics environment for collaborative mission evaluation. Two of these systems are critical for astronaut training. Crew must certify on SAFER and go through the Charlotte Mass Handling training prior to flying to the International Space Station (ISS). Typically, they also complete at least one collaborative visualization session to review any planned Extra Vehicular Activities (EVAs), or spacewalks, before an assigned flight. Given the volatility of new technologies, the graphics and simulation environments used are maintained to be hardware agnostic to preserve a high level of fidelity. Utilizing VR for astronaut training has proved to be effective and essential for these specific systems.

Angelica D. Garcia↗

NASA NTRSDate not supplied

GL4U: Bioinformatics training for students and educators using space omics data

NASA’s GeneLab project provides researchers open access to space-relevant experiment multi-omics data that can be mined to understand the effects of spaceflight on biological systems. To maximize the number of scientists who understand and utilize GeneLab data and data processing pipelines, GeneLab has created GeneLab for Colleges and Universities (GL4U). GL4U provides space biology-relevant training in bioinformatics to the next generation of scientists through direct and indirect approaches. The GeneLab team plans to host two annual data processing bootcamps, one for college-level students (direct) and one for college educators (indirect – training of trainers), in which participants learn to analyze GeneLab’s space-relevant omics data. The GL4U direct training pilot program was conducted in June 2021. During the pilot, students participated in a week-long bootcamp consisting of space biology-specific lectures and hands-on instruction using Jupyter Notebooks to analyze RNA sequence data. This pilot demonstrated the capacity of GL4U for training young scientists and encouraging data re-use. During the educator pilot, scheduled for June 2022, educators will receive materials and training to enable them to run the bootcamp at their home institutions or alternatively to adapt the content to implement within existing courses, thereby extending the reach of this initiative.

Amanda Marie Saravia-butler↗

NASA NTRSDate not supplied

GL4U: Bioinformatics Training for Students and Educators Using Space Omics Data

Amanda Marie Saravia-butler↗

NASA NTRSDate not supplied

Volcanic Field Sites for Artemis Testing and Training

The Artemis Program will reestablish human presence on the Moon and lead to a new era of scientific discovery and exploration. Led by the National Aeronautics and Space Administration (NASA), the Artemis Program is a collaboration of space agencies and companies around the world. An integrated effort between various disciplines of science, engineering, and mission operations is currently developing methods, facilities, and analog field locations to train astronauts and test hardware and concepts of operations. These efforts aim to best prepare for the next steps of human exploration on the lunar surface and beyond. Numerous terrestrial volcanic field sites were evaluated and selected for their unique roles in helping to prepare for the lunar surface mission phases. This effort heavily leveraged the comprehensive academic research conducted at these field sites, as well as the tremendous Apollo heritage. The currently selected volcanic field sites include the San Francisco Volcanic Field in Northern Arizona, the Potrillo Volcanic Field in southern New Mexico, the highlands of Iceland, and the Southwestern Nevada Volcanic Field. Within each of these volcanic field sites numerous specific testing and training locations are being further developed utilizing the analogous terrain and unique features in these regions. Recent Artemis testing and training events have been conducted at a number of these volcanic field sites by both a dedicated Artemis Geology Training Team and a Joint Extra Vehicular Activity (EVA) Testing Team. This presentation will highlight the selected sites as well as the objectives and accomplishments of some of the recent field-testing events and training courses. Additionally, we continually strive to pursue additional sites, locations, data sets, collaborations, and partnerships in this endeavor and welcome knowledge transfer and community input.

Trevor Graff↗

DOE OSTI2023

Systems and methods for active learning from sparse training data

A method for active learning using sparse training data can include training a machine learning model using less than ten first training data points to generate a candidate machine learning model. The method can include performing a Monte Carlo process to sample one or more first outputs of the candidate machine learning model. The method can include testing the one or more first outputs to determine if each of the one or more first outputs satisfy a respective convergence condition. The method can include, responsive to at least one first output not satisfying the respective convergence condition, training the candidate machine learning model using at least one second training data point corresponding to the at least one first output. The method can include, responsive to the one or more first outputs each satisfying the respective convergence condition, outputting the candidate machine learning model.

Sankaranarayanan, Subramanian↗

DOE OSTI2017

Training and certification of human organizational factors designers in nuclear plant engineering - 344

Performing an increasing number of small and large-scope nuclear Human Factors Programs, AREVA NP is facing the challenging situation that (1) well-trained experts in Human Factors or Usability Engineering are currently hard to find and (2) newly hired ones need at least two years 'on the job' to become familiar with the nuclear field and the complexity of engineering Human Factors in this field. To innovatively meet this challenge we focus instead on company-internal engineers. Demonstrated nuclear engineering knowledge and skills and pronounced interest to extend both renders them optimal applicants to the AREVA NP two years Human Organizational Factors Training and Certification program. The AREVA NP training shall convey Human Factors knowledge and skills within the limited time of two years. We also stress the need to change participants' mindset from a technically-biased to a human-centered approach, committed to involve end-users throughout design. Next to psychology principles, user testing, user-centered design, plant operation principles, and the acquisition and documentation of users' requirements, design constraints, and inherent tradeoffs, participants will be trained to apply learned methods in a real environment and meet the challenges of real projects. The following sections provide insights in the training conception and preparation phase, the identification of appropriate candidates, the development of course syllabuses, and the schedule of lectures and tests. The promising experiences made thus far are reported as well. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2022

CFD-trained ANN Model for Approximating Near-occupant Condition in Real-time Simulations

The main drawback of Computational Fluid Dynamics (CFD) simulations has been the time and resource consuming nature which is not suitable for real-time applications. In this work, we first generated numerous CFD models of a given indoor space to obtain airspeed, temperature, and mean radiant temperature near an occupant as training data. Several artificial neural networks (ANN) models were trained using this CFD simulated data to approximate near real-time environmental conditions for a given occupant. This trained ANN model approach is a part of a real-time simulation of building operations using a combination of software and real hardware (HVAC equipment) approaches. The preliminary results suggest that the CFD- generated training data and the trained ANN model can accurately approximate such conditions in a real-time application, a method that has great potential in building simulation and building digital twin areas of research.

Zhang, Yun↗

DOE OSTI2025

A geometric framework for momentum-based optimizers for low-rank training

Low-rank pre-training and fine-tuning have recently emerged as promising techniques for reducing the computational and storage costs of large neural networks. Training low-rank parameterizations typically relies on conventional optimizers such as heavy ball momentum methods or Adam. In this work, we identify and analyze potential difficulties that these training methods encounter when used to train low-rank parameterizations of weights. In particular, we show that classical momentum methods can struggle to converge to a local optimum due to the geometry of the underlying optimization landscape. To address this, we introduce novel training strategies derived from dynamical low-rank approximation, which explicitly account for the underlying geometric structure. Our approach leverages and combines tools from dynamical low-rank approximation and momentum-based optimization to design optimizers that respect the intrinsic geometry of the parameter space. We validate our methods through numerical experiments, demonstrating faster convergence, and stronger validation metrics at given parameter budgets.

Schotthoefer, Steffen [ORNL] (ORCID:00000002156965↗

DOE OSTI2023

Nuclear Criticality Safety Training: Needs and Efforts [Slides]

This presentation on Nuclear Criticality Safety Training: Needs and Efforts covers several points. This presentation starts with a quick overview of Nuclear Criticality Safety (NCS) Training. Then it lists current and ongoing training with a focus on the training in progress. This presentation then concludes with a look at supporting efforts.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

DOE OSTI2024

Importance of Hands-on Training within the Nuclear Industry

Almost all work performed within the nuclear industry requires some form of training, regardless of an individual’s education. While many of the hazards that exist in nuclear operations are present in other industries (e.g., industrial safety), nuclear operations have the unique hazards of radiation and inadvertent criticality. Therefore, these topics require additional training for individuals performing hands-on work or accessing those work areas. Many methods are available to train people including lectures (either with an instructor or automated video), reading procedures/policies, hands-on demonstrations, and on-the-job training.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2022

Pre-trained network-based transfer learning: A small-sample machine learning approach to nuclear power plant classification problem

Some research topics belonging to classification problems in the nuclear industry, such as fault diagnosis and accident identification, can be solved by feature extraction and subsequent application of statistical machine learning classifiers. Recently, deep neural network-based methods with automatic feature extraction and high accuracy have gained wide attention. They usually require large-scale training data, however, plant fault or accident data are scarce or difficult to obtain. Here this paper proposes a convolutional network (CNN)-based transfer learning method to solve this problem. The network's shallow layer is derived from a pre-trained CNN based on the ImageNet database to automatically extract features, and the deep layer is customized to match the classification problem. Data in non-image formats are converted to image formats and subsequently used to train the network. Case studies of rotating machines fault diagnosis show that the proposed method requires only limited training data to achieve high accuracy.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗